Cloud Computing Lecture #1 Parallel and Distributed Computing Jimmy Lin The iSchool University of Maryland Monday, January 28, 2008 This work is licensed.

Cloud Computing Lecture #1

Parallel and Distributed Computing

Jimmy LinThe iSchoolUniversity of Maryland

Monday, January 28, 2008

This work is licensed under a Creative Commons Attribution-Noncommercial-Share Alike 3.0 United StatesSee http://creativecommons.org/licenses/by-nc-sa/3.0/us/ for details

Material adapted from slides by Christophe Bisciglia, Aaron Kimball, & Sierra Michels-Slettvet, Google Distributed Computing Seminar, 2007 (licensed under Creation Commons Attribution 3.0 License)

iSchool

Today’s Topics Course overview

Introduction to parallel and distributed processing

iSchool

What’s the course about? Integration of research and teaching

Team leaders get help on a tough problem Team members gain valuable experience

Criteria for success: at the end of the course Each team will have a publishable result Each team will have a paper suitable for submission to

an appropriate conference/journal

Along the way: Build a community of hadoopers at Maryland Generate lots of publicity Have lots of fun!

iSchool

Hadoop Zen Don’t get frustrated (take a deep breath)…

This is bleeding edge technology

Be patient… This is the first time I’ve taught this course

Be flexible… Lots of uncertainty along the way

Be constructive… Tell me how I can make everyone’s experience better

iSchool

Things to go over… Course schedule

Course objectives

Assignments and deliverables

Evaluation

iSchool

My Role To hack alongside everyone

To substantively contribute ideas where appropriate

To serve as a facilitator and a resource

To make sure everything runs smoothly

iSchool

Outline Web-Scale Problems

Parallel vs. Distributed Computing

Flynn's Taxonomy

Programming Patterns

iSchool

Web-Scale Problems Characteristics:

Lots of data Lots of crunching (not necessarily complex itself)

Examples: Obviously, problems involving the Web Empirical and data-driven research (e.g., in HLT) “Post-genomics era” in life sciences High-quality animation The serious hobbyist

iSchool

It all boils down to… Divide-and-conquer

Throwing more hardware at the problem

Simple to understand… a lifetime to master…

iSchool

Parallel vs. Distributed Parallel computing generally means:

Vector processing of data Multiple CPUs in a single computer

Distributed computing generally means: Multiple CPUs across many computers

iSchool

Flynn’s Taxonomy

Instructions

Single (SI) Multiple (MI)D

ata

Mu

ltip

le (

MD

)SISD

Single-threaded process

MISD

Pipeline architecture

SIMD

Vector Processing

MIMD

Multi-threaded Programming

Sin

gle

(S

D)

iSchool

SISD

D D D D D D D

Processor

Instructions

iSchool

SIMD

D0

Processor

Instructions

D0D0 D0 D0 D0

D1

D2

D3

D4

…

Dn

D1

D2

D3

D4

…

Dn

D1

D2

D3

D4

…

Dn

D1

D2

D3

D4

…

Dn

D1

D2

D3

D4

…

Dn

D1

D2

D3

D4

…

Dn

D1

D2

D3

D4

…

Dn

D0

iSchool

MIMD

D D D D D D D

Processor

Instructions

D D D D D D D

Processor

Instructions

iSchool

Parallel vs. Distributed

SharedMemory

Parallel: Multiple CPUs within a shared memory machine

Distributed: Multiple machines with own memory connected over a network

Ne

two

rk c

on

ne

ctio

nfo

r d

ata

tra

nsf

er

D D D D D D D

Processor

Instructions

D D D D D D D

Processor

Instructions

iSchool

Divide and Conquer

“Work”

w1 w2 w3

r1 r2 r3

“Result”

“worker” “worker” “worker”

Partition

Combine

iSchool

Different Workers Different threads in the same core

Different cores in the same CPU

Different CPUs in a multi-processor system

Different machines in a distributed system

iSchool

Parallelization Problems How do we assign work units to workers?

What if we have more work units than workers?

What if workers need to share partial results?

How do we aggregate partial results?

How do we know all the workers have finished?

What if workers die?

What is the common theme of all of these problems?

iSchool

General Theme? Parallelization problems arise from:

Communication between workers Access to shared resources (e.g., data)

Thus, we need a synchronization system!

This is tricky: Finding bugs is hard Solving bugs is even harder

iSchool

Multi-Threaded Programming Difficult because

Don’t know the order in which threads run Don’t know when threads interrupt each other

Thus, we need: Semaphores (lock, unlock) Conditional variables (wait, notify, broadcast) Barriers

Still, lots of problems: Deadlock, livelock Race conditions ...

Moral of the story: be careful!

iSchool

Patterns for Parallelism Several programming methodologies exist to

build parallelism into programs

Here are some…

iSchool

Master/Workers The master initially owns all data

The master creates workers and assigns tasks

The master waits for workers to report back

workers

master

iSchool

Producer/Consumer Flow Producers create work items

Consumers process them

Can be daisy-chained

CP

P

P

C

C

CP

P

P

C

C

iSchool

All available consumers should be available to process data from any producer

Work queues divorce 1:1 relationship from producers to consumers

Work Queues

CP

P

P

C

C

shared queue

iSchool

And finally… The above solutions represent general patterns

In reality: Lots of one-off solutions, custom code Burden on the programmer to manage everything

Can we push the complexity onto the system? MapReduce…for next time